32 Research Report

Water Scarcity Variations within a Country: A Case Study of Sri Lanka

Upali A. Amarasinghe Lal Mutuwatta and R. Sakthivadivel

INTERNATIONAL WATER MANAGEMENT INSTITUTE P O Box 2075, Colombo, Sri Lanka Tel (94-1) 867404 • Fax (94-1) 866854 • E-mail [email protected] Internet Home Page http://www.cgiar.org/iwmi ISBN 92-9090-383-X ISSN 1026-0862 International Water Management Institute Research Reports

IWMI’s mission is to contribute to food security and poverty eradication by fostering sustainable increases in the productivity of water through better management of irri- gation and other water uses in river basins. In serving this mission, IWMI concentrates on the integration of policies, technologies and management systems to achieve work- able solutions to real problems—practical, relevant results in the field of irrigation and water resources. The publications in this series cover a wide range of subjects—from computer modeling to experience with water user associations—and vary in content from directly applicable research to more basic studies, on which applied work ultimately depends. Some research reports are narrowly focused, analytical, and detailed empirical stud- ies; others are wide-ranging and synthetic overviews of generic problems. Although most of the reports are published by IWMI staff and their collaborators, we welcome contributions from others. Each report is reviewed internally by IWMI’s own staff and Fellows, and by external reviewers. The reports are published and dis- tributed both in hard copy and electronically (http://www.cgiar.org/iwmi) and where pos- sible all data and analyses will be available as separate downloadable files. Reports may be copied freely and cited with due acknowledgment. Research Report 32

Water Scarcity Variations within a Country: A Case Study of Sri Lanka

Upali A. Amarasinghe Lal Mutuwatta, and R. Sakthivadivel

International Water Management Institute P O Box 2075, Colombo, Sri Lanka

i The authors: U. A. Amarasinghe is Research Statistician, L. Mutuwatta is Research Officer/GIS Analyst, and R. Sakthivadivel is Senior Irrigation Specialist, all of the International Water Management Institute.

The authors wish to acknowledge the valuable comments and suggestions of Dr. David Molden resulting in the improvement of the report. They also thank Drs. D. Merry, C. Panabokke, C. Perry, M. Samad, and D. Seckler for their comments at various stages of the preparation of the report and Mr. Sarath Gunasinghe for preparing the figures.

This work was undertaken with funds specifically allocated to IWMI’s Performance and Impact Assessment Program by the European Union and Japan, and from allocations from the unrestricted support provided by the Governments of Australia, Canada, China, Denmark, France, Germany, Netherlands, and the United States of America; the Ford Foundation; and the World Bank.

Amarasinghe, U. A., L. Mutuwatta, and R. Sakthivadivel. 1999. Water scarcity variations within a country: A case study of Sri Lanka. Research Report 32. Colombo, Sri Lanka: International Water Management Institute.

/ water scarcity / case studies / irrigated farming / rice / indicators / water resources / water supply / demand / runoff / river basins / irrigation requirements / Sri Lanka /

ISBN 92-9090-383-X ISSN 1026-0862

© IWMI, 1999. All rights reserved.

Responsibility for the contents of this publication rests with the author.

The International Irrigation Management Institute, one of sixteen centers supported by the Consultative Group on International Agricultural Research (CGIAR), was incorporated by an Act of Parliament in Sri Lanka. The Act is currently under amendment to read as International Water Management Institute (IWMI).

ii Contents

Summary v

Introduction 1

Water-Scarcity Indicators 3

Water Supply and Demand 4 Domain of Analysis 4 Water Resources 4 Water Withdrawals 10 Demand Projections 13

Water-Scarcity Indicators: District-Wise Distribution 15 Falkenmark Indicator 17 UN Indicator 17 IWMI Indicator 17 Comparison of the Three Indicators 21

Conclusion 21

Appendix A. Runoff Estimates of River Basins in Sri Lanka 23

Appendix B. Irrigation Requirement 26

Literature Cited 28

iii Summary

Several studies on present and future water According to most scarcity criteria, the national scarcities rank Sri Lanka as a country with either statistics indeed show no serious present or future little or no water scarcity or moderate water- water scarcity. However, a different picture scarcity conditions. None of these studies have emerges at the district level. Five districts (25% of considered the spatial and temporal variations of the total land area) in the maha season, and nine water availability and demand within the country. districts (43% of the total land area) in the yala This report examines the variations of water supply season withdrew more than 50 percent of their and demand and the differences of water scarcities water resources in 1991. These districts already between different districts in Sri Lanka under both have absolute water-scarce conditions according to present conditions and projected conditions in some criteria. 2025. A few more districts will enter into the absolute The results of this study indicate wide water-scarce category in 2025 under scenario 1. temporal and spatial variations of available water However, if the irrigation sector efficiency can be resources and demand. The total utilizable water doubled by 2025, only four districts in the maha resources per unit area in districts range from 0.03 season and nine districts in the yala season will meter to 1.43 meters in the maha (wet) season have severe water-scarce conditions. Though these and from 0.02 meter to 1.7 meters in the yala (dry) districts are identified as being severely water- season. On the demand side, the dry zone scarce, they can meet their 2025 demand by accounted for more than 90 percent of the current withdrawing at or below the current withdrawal withdrawals, whereas only 44 percent of the levels. population lived there in 1991. The heavy Whether the country has the institutional and withdrawal in the dry zone was mainly due to the financial capacity to attain the high irrigation higher share of irrigation demand. efficiency needed in the second scenario is not Water demand for the year 2025 is projected clear. However, at the current level of irrigation under two irrigation scenarios. The first scenario efficiency, the majority of the districts in the dry assumes the same irrigation sector efficiency, i.e., zone—Ampara, Anuradhapura, Batticaloa, the ratio of irrigation requirement to primary Hambantota, Jaffna, Killinochchi, Kurunegala, irrigation withdrawal in 2025 at the current level. Mullaitiv, Polonnaruwa, Puttalam, Trincomalee, and The second scenario assumes increased irrigation Vavunia—will face severe seasonal or year-round efficiency in 2025. Domestic and industrial absolute water-scarce conditions. These districts in demands are projected at the level of basic human the dry zone share more than 75 percent of the needs or at the current level of withdrawals, irrigation withdrawals at present and also the whichever is higher. Demand projections for 2025 highest increase in withdrawals projected for the show that the dry zone will again account for more future. Therefore, the water scarcities in the dry than 90 percent of the total water withdrawals. zone will have a severe impact on the country’s With an increased irrigation efficiency scenario, the future food production. total withdrawal demand in the country (and Also, contrary to common belief, Galle district especially in the dry zone) can be reduced by in the wet zone is also identified as having severe almost half. water-scarce conditions in an economic sense.

v Though this district has sufficient water resources, variations. When the same indicators are used at it will have to at least double its withdrawal to subunit level, a substantial area of the country meet the 2025 demand. comes under severe water-scarce conditions. The present study clearly illustrates that the A knowledge of subunit level water scarcities is statistics in the form of aggregated information at very important because most of the food national level sometimes mask issues of local requirement of the country at present comes from water scarcity. This is especially true when vast water-scarce regions and projected additional spatial and seasonal variations of water supply requirements are also to be met by the same and demand are present. Sri Lanka, even though a regions. small country, is a good example with such

vi Water Scarcity Variations within a Country: A Case Study of Sri Lanka

Upali A. Amarasinghe, Lal Mutuwatta, and R. Sakthivadivel

Introduction

Several studies on present and future water season rainfall is received from October to scarcities rank Sri Lanka as a country with either December. little or no water-scarcity or moderate water- On the demand side, the agriculture sector scarcity conditions (Falkenmark, Lundqvist, and used 96 percent of the total withdrawals in 1991 Widstrand 1989; Engleman and Leroy 1993; (ESCAP 1995). This was mainly for rice irrigation. Raskin et al. 1997; Seckler et al. 1998; Seckler, More than 85 percent of the 1991 irrigated rice Barker, and Amarasinghe 1999). These studies area was in the dry zone, which has the most used the aggregated information presented in the demand for water; on the other hand, the water form of statistics at national level. However, the availability in the wet zone is higher. If water is aggregated statistics on water scarcity at national scarce in regions where most of the withdrawals level are sometimes misleading for countries with are for irrigation, the impact on future food large regional variations. All these earlier studies production will be very significant for the following had ignored spatial and temporal variations of reasons. water availability and demand from their water- First, Sri Lanka withdrew a mere 31 liters/ scarcity calculations. The main objective of this person/day in 1991 for the domestic and industrial study is to assess the spatial and seasonal sectors (ESCAP 1995). This level of supply is variations of water supply and demand, and also even below the minimum level for the basic water scarcities at district level in Sri Lanka. requirements for domestic purposes (Gleick 1996). Sri Lanka experiences high seasonal and The changing socioeconomic conditions will spatial variations in rainfall due to the bimonsoonal demand more water for domestic purposes. Along climatic pattern (northeast monsoon from October with this, the rapid growth in the industrial sector to March and southwest monsoon from April to will also demand more water. September). It has two seasons: maha season Second, the greater part of the rice-irrigated from October to March and yala season from April area is in the dry zone. Recent trends show that to September. Rainfall patterns divide the country the rice area under rain-fed conditions has not primarily into two climatic zones: the wet zone recorded any increase (Aluwihare and Kikuchi and the dry zone. The wet zone, comprising one 1991). In fact, a decreasing trend in rain-fed area fourth of the land area receives an average of cultivation has been seen after the late 1970s. 2,350 mm of annual rainfall distributed in the two Also, no significant increase in yield has been seasons. The rest of the area, called the dry seen in either rain-fed or irrigated rice during zone, receives an average of 1,450 mm of annual recent years. All indications are that the yield per rainfall. In the dry zone, more than two-thirds of unit of land has reached a plateau since the the annual rainfall is received during the maha 1980s (Wijayaratna and Hemakeerthi 1992). season, and more than 70 percent of the maha Therefore, if the current rates of growth in rice

1 area and in rice yield are indications for future the long run the actual value will exceed the development, production increases in rice may assessed value every other year. On the other have to come totally from increasing the gross hand, the use of the 25th percentile (75 percent irrigated area. However, the greater part of the exceedence probability rainfall) will imply that irrigated rice area is in the dry zone, and the water actual runoff in the long run will exceed the availability there is less than that in other areas. estimated runoff in 3 out of every 4 years. Additionally, competition from domestic and Moreover, 75 percent exceedence probability industrial sectors will also increase. Therefore, rainfall is used for river basin planning studies. water scarcities in the dry zone will have a severe Therefore, we have selected the 75 percent impact on meeting the additional food exceedence probability rainfall for assessing requirements of the increasing population. utilizable river runoff. In this respect, it is of vital importance to Future water demands at district level are understand the regional scarcities of water and studied under two scenarios: The first scenario is their impact on future food production. Therefore, that the irrigation sector efficiency will be the the main objectives of this report are to assess: same in the future as at present. The second scenario assumes higher irrigation efficiency than • the differences in the present and future at present. These scenarios will be discussed in situations of water supply and demand at more detail in the respective sections. district level Of course, there are certain limitations to our analysis. Our estimates of current water • the existence of water scarcity at present and availability and demand for a unit may slightly their differences at district level differ from the actual value of the unit. This is mainly due to the difficulty in obtaining reliable • the district-level differences of future water data on water supply and demand at district level. scarcities assuming that the additional rice However, we believe that our method of estimation production required in the future would have to would enable us to compare the present and be met totally from irrigated agriculture future water supply and demand between districts in a consistent manner. This would also give In most studies in Sri Lanka, river enough information to illustrate the magnitude of discharges to the sea (runoff) are assessed water scarcity at district level. under average rainfall conditions. Such an For our analysis, we consider 1991 as the assessment is substantially influenced by the base year to assess the current status, and the extreme rainfall years. In fact, this is the case in year 2025 for future projections. In the next Sri Lanka where seasonal rainfall distribution is section, three water scarcity criteria described in skewed to the right due to years with extremely the literature are briefly introduced. In section 3, high rainfall. The use of appropriate percentiles the present and future water supply and demand of the rainfall distribution will overcome the situations are discussed. The differences of water undue influence of years of extreme rainfall in scarcity indicators at district level and the averaging. What percentile of the rainfall distribution of population, land area, and irrigated distribution is appropriate is the next question. If area for different scarcity categories are discussed one uses the median or the 50th percentile (50 in section 4. The concluding remarks are in percent exceedence probability rainfall), then in section 5.

2 Water Scarcity Indicators

Falkenmark, Lundqvist, and Widstrand (1989) have withdrawals from 20 to 40 percent are considered to used annual per capita water availability to define have medium to severe water scarcity. Countries in water scarcity thresholds. In their criterion of water this range are required to take effective steps to scarcity, a country is considered water-scarce if manage their water supply and demand in a way the per capita annual water supply falls below that the withdrawals for different sectors are 1,700 cubic meters (m3). Above this level of per sustainable. Countries with withdrawals from 10 to capita water supply, water scarcities are 20 percent are considered to have moderate water considered to be rare and, if they exist, they are scarcity, while those with less than 10 percent only problems within a few localities. Below the withdrawals are considered to have little or no water 1,700 m3 level, a country faces seasonal or scarcity. Hereafter this will be referred to as the UN regular water-stressed conditions. If the annual per indicator. capita water supply falls below 1,000 m3, water In a study by the International Water shortages begin to hamper the health and well- Management Institute (IWMI) (Seckler et al. 1998; being of human beings, and if it falls below 500 Seckler, Barker, and Amarasinghe 1999), water m3, shortages are severe constraints to human scarcities are defined in terms of two factors: The life. Hereafter we refer to this as the Falkenmark first is the future withdrawal as a percent of the indicator. We call the four categories as severe, available water resources. The second is the medium to severe, moderate, and little or no water increase in additional water withdrawal needs. A scarcity (table 1). country is defined to be absolutely water-scarce if In a recent UN study (Raskin et al. 1997), water the demand is more than 50 percent of the available scarcity thresholds were defined in terms of the water resources. Scarcities of other countries are percent of water resources withdrawn for different ranked according to their future development needs. uses. A country is considered to be severely water- For example, a country is severely water-scarce in scarce if the withdrawals exceed 40 percent of the an economic sense, if the future demand is more total supply. Above this level, countries will have to than double the present withdrawal level. In this depend more and more on desalination or on the report, absolute water scarcity is called severe use of groundwater in an unsustainable manner. In water-scarce category, while the severe economic this category, the water scarcity is the limiting water-scarce category is called medium to severe- factor to economic growth. Countries with water scarce category (table 1).

TABLE 1. The water scarcity levels of the three indicators.

Indicator and criteria Falkenmark UN IWMI

Scarcity level IF = Annual per capita IUN = Withdrawal as a % IIWMI1 = Withdrawal as a % water supply of water supply of water supply

IIWMI2 = Future withdrawals as a % of current withdrawals

3 Severe IF < 500m IUN >40% IIWMI1 > 50% 3 3 Medium to severe 500m < IF < 1,000m 20% < IUN <40% IIWMI1 <50% & IIWMI2 >200% 3 3 Moderate 1,000m < IF <1,700m 10% < IUN <20% IIWMI1 <50% & 125% < IIWMI2< 200% 3 Little or no 1,700m < IF IUN <10% IIWMI1 <50% & IIWMI2 <125%

3 Water Supply and Demand

Domain of Analysis Water Resources

An ideal domain for conducting water resources The water resources for a district consist of the analysis is a hydrological unit such as a river following components (Seckler et al. 1998): basin. However, in this report the administrative district is considered as the basic unit of analysis. • surface inflow to a district generated from the For several reasons, most important of which is precipitation within the district and inflow from that population data, some water supply data, and neighboring districts water-use-related data (which are very important to • surface outflow, i.e., that portion of surface our study) are available at administrative unit inflow getting out of the district level. Also, from an analysis at district level, a regional picture at larger administrative units such • change in storage (net change in reservoir as the provincial or at climatic zone levels can be storage is accounted for, while net change in obtained by summation. groundwater storage is assumed to be negligible) In Sri Lanka, several administrative units share the water of major river basins. For example, the Then the utilized flow is the sum of the Mahaweli basin (figure 1) covers about 18 percent surface inflow and the change in storage minus of the land area, and contributes to the water the surface outflow. Most of the above compo- resources of several administrative districts in both nents at district level are not measured, and need the wet zone and the dry zone. Likewise, the Kalu to be estimated. For example, the inflows and the ganga and Kelani ganga basins, which cover more outflows from one district to another are not than 8 percent of the land area, lie completely recorded. Therefore, a different methodology has inside the wet zone benefiting from the rainfall from to be adopted for computing the net inflows and several administrative districts. At present, there surface runoff. The suggested methodology is are no apparent major conflicts among administra- based on the computation of utilizable seasonal tive districts in sharing the water resources of river runoff of a basin and apportioning it among basins. However, a changing socioeconomic and districts contributing to that discharge, proportional political environment will increase the competition to the area of contribution. As discussed in the among districts for water resources of these introductory section, the potentially utilizable flow basins. Therefore, it is important to have informa- in this study is based on the 75 percent tion on the current supply and demand situation of exceedence probability rainfall, and not on the water and also of the water scarcities (if they exist) average rainfall. The following are the steps at district level and also at larger administrative unit involved in the computation: levels such as the provincial level. There are 25 administrative districts in Sri • Identify basins having a fairly long period of Lanka and in 1991 its total population was 17 record of discharge and rainfall, and select a million (table 2). The wet zone contains 9 districts, period where the discharges are not affected mainly in the western, central, and southern unduly by storage structures within the basin. provinces. These districts, with only 25 percent of the land area, contain 56 percent of the total • For the period selected, convert values of population. The remaining districts are considered monthly rainfall and discharges into values of to be in the dry zone. seasonal rainfall and discharges.

4 FIGURE 1. Major river basins in Sri Lanka.

3 River basin Jaffna

River basin boundary

District boundary Kilinochchi

Mullativu

Mannar Vavuniya Aruvi Aru

Trincomalee

Yan Oya

10 Anuradhapura 9 Polonnaruwa

Puttalam anga G

Batticaloa weli Deduru Oya 8

Maha

Matale

7 Oya Maha Kandy Gal Oya 1

Gampaha 2 Kegalle Nuwara Eliya Ampara Kelani Ganga Badulla 6 Colombo

Ganga Monaragala Kalu 5 Menik

Ratnapura Ganga 3 Kalutara 4

W alaw

e G Hambantota anga Galle 0204060 Matara KILOMETERS

5 • Develop seasonal rainfall-runoff relationships. • Superimpose the district boundary map on the river basin map and identify districts • Using 75 percent exceedence probability and their areas that constitute each basin. seasonal rainfall, determine basin seasonal runoff from the rainfall-runoff relationships. • Seasonal runoff of a basin is apportioned to each district proportionally to the 75 percent • For basins where seasonal rainfall-runoff exceedence probability total rainfall received relationships cannot be established, the runoff on the area of each district that intersects the is computed by linearly interpolating the runoff basin. per unit area of adjoining basins where rainfall- runoff relationships exist.

TABLE 2. Land area, population, per capita domestic and industrial withdrawals, and storage capacity.

1991 Population 1991 1991 Industrial Potential Land Average Total Urban With Per capita Output Withdrawal storage Unit area rainfall as a % of piped per day per capita capacity total water domestic per day supply withdrawal 1,000km2 mm million % million liters billionRS liters km3 C11 C2 C31 C4 C51 C6 C71 C8 C91 Sri Lanka 65.6 1,672 17.3 21 4.6 31 112 31 6.05 Wet Zone (WZ) 23.3% 2,353 56.0% 28 64% 35 89% 49 26% Dry Zone (DZ) 76.7% 1,465 44.0% 13 36% 26 11% 8 74% Districts-WZ Colombo 1.1% 2,528 11.4% 74 24% 65 48% 132 0% Galle 2.5% 2,344 5.5% 21 5% 31 2% 9 0% Gampaha 2.4% 2,210 8.9% 28 10% 35 30% 105 0% Kalutara 2.1% 2,655 5.5% 21 6% 31 1% 4 0% Kandy 3.0% 2,155 7.3% 13 6% 26 1% 4 21% Kegalle 2.6% 2,388 4.4% 8 3% 22 1% 10 0% Matara 2.0% 2,217 4.5% 11 4% 25 1% 7 0% Nuwara Eliya 2.7% 2,122 3.1% 7 2% 22 3% 33 0% Ratnapura 5.0% 2,535 5.5% 7 4% 22 2% 9 5% Districts-DZ Ampara 6.7% 1,509 2.8% 14 2% 26 1% 6 26% Anuradhapura 10.9% 1,305 4.1% 7 3% 22 0% 0 13% Badulla 4.4% 1,776 4.2% 8 3% 23 1% 7 1% Batticaloa 4.4% 1,574 2.4% 24 3% 33 1% 7 3% Hambantota 4.0% 1,691 3.0% 10 2% 24 0% 1 2% Jaffna(1) 1.6% 1,104 5.0% 33 6% 38 1% 3 0% Killinochchi 1.9% 1,102 0.6% 9 0% 23 1% 27 3% Kurunegala 7.3% 1,654 8.3% 4 5% 20 1% 3 4% Mannar 3.0% 1,072 0.8% 14 1% 26 1% 21 2% Matale 3.0% 1,690 2.4% 11 2% 24 0% 5 1% Monaragala 8.6% 1,587 2.0% 2 1% 19 3% 40 1% Mullaitivu 4.0% 1,145 0.5% 9 0% 23 1% 30 0% Polonnaruwa 5.0% 1,532 1.8% 8 1% 22 0% 0 8% Puttalam 4.7% 1,478 3.5% 13 3% 25 2% 17 2% Trincomalee 4.2% 1,426 1.8% 32 2% 38 1% 9 4% Vavuniya 3.0% 1,172 0.7% 19 1% 30 1% 24 3% 1Wet-zone, dry-zone, and district values are given as a percent of the Sri Lankan total.

6 • Net diversions from outside the district and probability rainfall is calculated first (see appendix potential carryover from one season to another A for more details). The contributions of districts were added to the computed runoff to to the total runoff are worked out next (table 3). estimate the total water supply available to The runoff figures indicate wide spatial and the district. seasonal variations. For example, though the total maha season runoff in the two climatic zones is Contribution to Surface Runoff similar, the yala season total runoff is significantly different. The maha season runoff per unit area in Based on the above methodology, the seasonal the dry zone is less than one-third of that in the runoff of basins at 75 percent exceedence wet zone (C2, table 3). The yala season runoff per

TABLE 3. Seasonal water resources. Maha season Yala season Contribution Carry- Net Water Contribution Carry- Net Water to runoff over divers- resources to runoff over diver- resources Unit Total Depth storage ions Total Depth on Total Depth storage sions Total Depth on land area land area km3 mkm3 km3 km3 mkm3 mkm3 km3 km3 m C11 C2 C41 C5 C61 C7 C81 C9 C111 C12 C131 C14 Sri Lanka 24.29 0.37 0.96 0.00 22.99 0.35 16.95 0.26 2.26 0.00 18.25 0.28 Wet zone (WZ) 51% 0.81 100% -1.15 49% 0.74 81% 0.90 43% -0.74 71% 0.85 Dry zone (DZ) 49% 0.24 0% 1.15 51% 0.23 19% 0.06 57% 0.74 29% 0.10 Districts-WZ Colombo 2% 0.77 3% 2% 0.77 4% 0.88 1% 3% 0.88 Galle 7% 0.99 0% 7% 0.99 11% 1.13 0% 10% 1.13 Gampaha 3% 0.53 0% 4% 0.53 5% 0.56 0% 5% 0.56 Kalutara 8% 1.43 0% 9% 1.43 14% 1.70 0% 13% 1.70 Kandy 4% 0.56 65% -0.57 2% 0.27 5% 0.40 28% -0.28 3% 0.25 Kegalle 7% 0.96 0% 7% 0.96 13% 1.27 0% 12% 1.27 Matara 4% 0.73 1% 4% 0.73 5% 0.67 0% 5% 0.67 Nuwara Eliya 5% 0.64 0% -0.58 2% 0.30 7% 0.71 0% -0.45 4% 0.45 Ratnapura 11% 0.82 31% 12% 0.82 18% 0.92 13% 17% 0.92 Districts-DZ 0.00 Ampara 7% 0.39 0% 5% 0.28 1% 0.03 21% 3% 0.13 Anuradhapura 3% 0.10 0% 0.27 3% 0.11 1% 0.02 10% 0.23 3% 0.09 Badulla 6% 0.49 0% 6% 0.48 3% 0.19 1% 3% 0.20 Batticaloa 4% 0.31 0% 4% 0.30 0% 0.02 2% 1% 0.04 Hambantota 3% 0.25 0% 3% 0.23 2% 0.15 2% 2% 0.17 Jaffna(1) 1% 0.15 0% 1% 0.15 0% 0.04 0% 0% 0.04 Killinochchi 0% 0.07 0% 0% 0.06 0% 0.01 1% 0% 0.03 Kurunegala 4% 0.20 0% 4% 0.19 3% 0.11 3% 3% 0.12 Mannar 0% 0.05 0% 0% 0.03 0% 0.01 1% 0% 0.03 Matale 3% 0.39 0% 0.03 3% 0.39 2% 0.17 1% 0.03 2% 0.19 Monaragala 7% 0.29 0% 7% 0.28 2% 0.06 0% 2% 0.06 Mullaitivu 1% 0.09 0% 1% 0.09 0% 0.02 0% 0% 0.02 Polonnaruwa 6% 0.41 0% 0.79 9% 0.61 2% 0.11 7% 0.45 5% 0.29 Puttalam 1% 0.07 0% 1% 0.06 1% 0.03 2% 1% 0.05 Trincomalee 3% 0.28 0% 0.06 3% 0.27 1% 0.05 3% 0.02 1% 0.08 Vavuniya 1% 0.10 0% 1% 0.08 0% 0.02 2% 1% 0.05 1Wet-zone, dry-zone, and district values are given as a percent of Sri Lankan total.

7 unit area in the dry zone is only about 6 percent storage from one season to another can easily be of that in the wet zone (C9, table 3). It is lower or higher due to high variations in seasonal important to note that there also exist some and annual rainfall. For example, a weak northeast substantial seasonal differences. For example, in monsoon and hence a lower rainfall in the maha the dry zone, the yala season runoff is about a season would reduce the maha season runoff and quarter of the maha season runoff (C8, table 3). hence the reservoir storage available for the yala season. Net carryover runoffs as a percent of Reservoir Storage storage capacity from one season to another are found in C4 and C11 of table 3. At present, Sri Lanka has an estimated 6 km3 of 1 potential storage capacity. This consists of 5.25 Net Diversions km3 from major reservoirs, 0.38 km3 from medium reservoirs, and 0.41 km3 from minor tanks.2 Under the Mahaweli Development Program, there The contributions from reservoir storage to are substantial amounts of water diverted from the seasonal water resources are computed as wet zone area of the Mahaweli basin to the dry described below. The storage for withdrawal for a zone basins. The study by Sakthivadivel et al. given season is available in two ways. One is the (1995) tabulated the Mahaweli diversions to the storage available at the beginning of the season. reservoirs in the Matale, Anuradhapura, The other is the inflow to the reservoirs during the Polonnaruwa, and Trincomalee districts in the dry season due to precipitation in the previous zone. The probable diversions at 75 percent season. In the wet zone, except in the Kandy probability level are given in C5 and C12 in table 3. district, we assume that runoff equivalent to 100 percent of the storage capacity is available from Potentially Utilizable Seasonal Water Resources one season to another. In the Kandy district we assume the availability as only 50 percent of the The potentially utilizable seasonal water resource storage capacity. In the dry zone, we assume that for a district is the aggregate of the contribution to the maha season runoff equivalent to 30 percent runoff from a district, carryover runoff as storage of the storage capacity is available for the yala from the previous season, and the net diversions season water resources, while no yala season to the district. The total water resources of the runoff is available as storage for the maha season two seasons show vast differences between water resources (table 4). Certainly, the districts (figure 2). Maha season water resources assumptions on availability of runoff as reservoir per unit area range from 1.43 meters in the Kalutara district in the wet zone to 0.03 meter in TABLE 4. Carryover runoff between seasons as a percent of storage the Mannar district in the dry zone. Yala season capacity. water resources per unit area (C14) range from 1.7 meters in the Kalutara district to 0.02 meter in Climatic Carryover runoff as a % of storage capacity the Mullaitivu district. Though the maha season zone Maha to yala season Yala to maha season total water resources are similar in the two Wet zone 100 100 climatic zones (C6, table 3), the yala season Dry zone 30 0 totals are significantly different (C13, table 3). Of

1The potential storage capacity was estimated from the figures supplied in various publications by Arumugam (1969); Irrigation Depart- ment (1975); and Mahaweli Authority of Sri Lanka (1995). 2Storage capacity of the minor tank category may be higher than the value reported here, as data for all the minor tanks were not available (Irrigation Department 1975).

8 AMPARA BATTICALOA MONARAGALA HAMBANTOTA TRINCOMALEE BADULLA POLONNARUWA MATALE KANDY NUWARA ELIYA NUWARA MATARA RATNAPURA VAVUNIYA MULLAITIVU ANURADHAPURA KEGALLE KILINOCHCHI GALLE KURUNEGALA MANNAR KALUTARA JAFFNA Yala season Yala GAMPAHA COLOMBO PUTTALAM < 0.1 0.1 - 0.3 0.3 - 0.6 0.6 - 0.9 > 0.9 Depth in meters AMPARA BATTICALOA MONARAGALA HAMBANTOTA TRINCOMALEE BADULLA POLONNARUWA MATALE KANDY NUWARA ELIYA NUWARA MATARA RATNAPURA VAVUNIYA MULLAITIVU ANURADHAPURA KEGALLE KILINOCHCHI GALLE KURUNEGALA MANNAR KALUTARA JAFFNA Maha season GAMPAHA COLOMBO PUTTALAM FIGURE 2. Seasonal water resources per unit area.

9 the total water resources in the yala season only The per capita industrial withdrawal in 1991 29 percent is available for the dry zone. was 31 liters/person/day (C8, table 2). As in the domestic sector, the distributions of withdrawals at district level are not available for this study. Water Withdrawals Therefore, the industrial withdrawal4 for a district in 1991 is assumed to be proportional to its 1991 Based on population growth, increase in domestic industrial output (C7, table 2). There are demand, and on growth in industry and irrigation, substantial spatial differences of per capita ESCAP (1995) estimated that Sri Lanka’s total industrial withdrawal. For example, the per capita water withdrawal in 1991 was 9.77 km3, of which industrial withdrawal in the wet zone is more than 96 percent was used by the agriculture sector and six times that in the dry zone. 2 percent each by the domestic and industrial sectors. Irrigation Withdrawals

Irrigated agriculture accounted for 96 percent of Domestic and Industrial Withdrawals in 1991 the 1991 water withdrawals, or 9.38 km3. The total The estimated total domestic withdrawal in 1991 irrigated area in 1991 was 642,000 hectares. Of was the same as for the total industrial withdrawal, these, an area of 575,000 hectares (348,000 which was 0.19 km3. The withdrawals to domestic hectares in the maha season, 227,000 hectares in sector at district level were not available for this the yala season) was under rice. About 10 percent study. However, it is known that 70 percent of the of the irrigated area was under other field crops urban and 15 percent of the rural population are (OFCs). There are wide variations of irrigated area served by piped water (ESCAP 1995). According to between regions. For example, the districts in the the last census (Department of Census and dry zone contain more than 85 percent of the total Statistics 1995), only 21 percent of Sri Lanka’s annual irrigated area. population lived in urban areas. Therefore, in the Seasonal irrigation withdrawal (IRR)5 for a absence of withdrawal data at district level, we given district is assumed to be proportional to the assumed that the domestic water withdrawal3 for seasonal irrigation requirements for the area each district is proportional to the urban and rural irrigated by the districts. The irrigation requirement population supplied with piped water. of the two seasons is the sum of net Sri Lanka withdrew only 31 liters/person/day evapotranspiration (NET) of the first 5 months in for domestic withdrawal in 1991 (C6, table 2). the maha season and first 4 months in the yala However, the estimates of variations between season. The net evapotranspiration for a month is districts are substantial. These range from a taken as the potential evapotranspiration (ETo) maximum of 65 to a minimum of 19 liters/person/ minus the effective rainfall. The effective rainfall day in the Colombo and Monaragala districts, for a district is calculated as follows: First, the respectively. difference between the average rainfall volume and

3 th The domestic withdrawal for i. district, DOM(i), is estimated as, , where p(i) is the percent of

th urban population and p(J) is the percent of the total population of the i. district, and DOM is the total domestic withdrawal. 4 th The industrial withdrawal for the i. district, IND(i) is estimated as IND (i) = ind(i) x IND, where ind(i) is the percent contribution from the th i. district to the total industrial output, and IND is the total industrial withdrawal.

5 th th The irrigation withdrawal IRR(i,j) for the i. district in the j. season is estimated as , where NET(i,j) is the total

th th irrigation requirement for the i. district in the j. season, and IRR is the total annual agriculture withdrawal.

10 the runoff volume generated from that rainfall is Irrigation sector efficiency tells us more about the calculated. Then 90 percent of this difference is degree of scarcity than project efficiency but is taken as the effective rainfall while the other 10 much harder to estimate. Values are often percent is lost as non-recoverable percolation. The available for efficiencies of various projects, but it calculation of NET for each district will be is rare to find values of sector efficiency. discussed in more detail in appendix B. Therefore, we need a way to convert average project efficiency to sector efficiency. multiplier (M) Irrigation Efficiency and Water Recycling Let us define the as the ratio of total withdrawals to primary water. That is: Recycling of water within basins in Sri Lanka, as in many locations worldwide, plays an important Total withdrawals = Primary water x M role in the understanding of water supply and demand. When water is withdrawn for irrigation, Using the relationships in 1 and 2, we find part of the water is consumed by that: evapotranspiration. Part of the water then flows out of the irrigation system either along the Sector efficiency = M x Project efficiency surface or through the groundwater system. If (MxProject efficiency cannot be greater than 1) another irrigation system uses the outflow again, we consider this as recycled water. This tells us that in order to increase sector In a basin or district, we define primary water efficiency, we must either improve the project as water withdrawals originating from runoff, or as efficiency or increase the multiplier. augmented flows to the area, but excluding Unfortunately, there is insufficient information recycled water. If there were no recycling of water, on recycling and sector efficiency. On the other the sum of all withdrawals would be equal to hand, we know that information on withdrawals primary water. With recycling, the sum of all includes recycling, and thus estimates of withdrawals is greater than the primary water. An efficiencies and water supplies are misleading. So, important point is that water consumption from based on our experience, we assume a multiplier withdrawals cannot exceed primary water. of 4/3, which implies that one-third of the primary Within an irrigation project, we define project water is recycled. efficiency as the ratio of NET to irrigation withdrawal, i.e., Primary Irrigation Withdrawals

Project efficiency = NET/Total withdrawals There are significant differences of primary irrigation withdrawals between districts. The dry The project efficiency within a district or basin zone district’s share of the total is 82 percent in is the ratio of NET to the sum of all withdrawals. the maha season and 92 percent in the yala We define irrigation sector efficiency as the ratio season. This is mainly due to the high share of of NET within a district or basin to primary water, irrigation withdrawals. In terms of per capita i.e., withdrawals, figure 3 shows the extent of the differences between districts. While some districts Sector efficiency = NET/Primary water in the wet zone have less than 50 m3 per capita water withdrawals, districts in the dry zone have If there is recycling, the irrigation sector more than 500m3, and some exceed even efficiency is higher than project efficiency. 1,000 m3 per capita water withdrawals.

11 AMPARA BATTICALOA MONARAGALA HAMBANTOTA TRINCOMALEE BADULLA POLONNARUWA MATALE KANDY NUWARA ELIYA NUWARA MATARA RATNAPURA VAVUNIYA MULLAITIVU ANURADHAPURA KEGALLE KILINOCHCHI GALLE KURUNEGALA MANNAR KALUTARA JAFFNA Yala season Yala GAMPAHA COLOMBO PUTTALAM < 50 50 - 250 250 - 500 500 - 1000 > 1000 Withdrawals in cubic meters AMPARA BATTICALOA MONARAGALA HAMBANTOTA TRINCOMALEE BADULLA POLONNARUWA MATALE KANDY NUWARA ELIYA NUWARA MATARA RATNAPURA VAVUNIYA MULLAITIVU ANURADHAPURA KEGALLE KILINOCHCHI GALLE KURUNEGALA MANNAR KALUTARA JAFFNA Maha season GAMPAHA COLOMBO PUTTALAM FIGURE 3. Per capita water withdrawals in 1991.

12 Demand Projections due to improved exogenous factors such as improved fertilizer, better technology, etc. Irrigation • The additional rice production in 2025 will We assume that one-third of the primary irrigation come totally from the irrigation sector. withdrawals will be recycled in 2025, i.e., a multiplier of 4/3. Irrigation demand in 2025 is Under the above assumptions, first we assessed under two scenarios. The first scenario calculate the required growth in rice irrigated area assumes the same irrigation sector efficiency in (table 5). The gross irrigated rice area in 1991 was 2025 as at present. In the second scenario, 69 percent of the total rice cultivated area (row 2, irrigation sector efficiency in 2025 is expected to table 5). The estimated yield was 3.84 tons/ double the current levels. hectare under irrigation, and 2.60 tons/hectare Rice is the major irrigated crop in Sri Lanka under rain-fed cultivation (Samad 1997). This (about 90% in each season). We project the shows that 77 percent of the total rice production growth of rice irrigated area first. For both in 1991 was from irrigation (row 5, table 5). scenarios, growth in rice irrigation for 2025 was According to the United Nations medium computed using the following assumptions: projections (UN 1995), the Sri Lankan population will increase by 45 percent during the period • The level of per capita rice production 1991–2025. Therefore, under the first assumption, (aggregate of irrigation and rain-fed) will stay the total rice demand in 2025 will be 4.187 million the same through the period 1991–2025. This metric tons (row 7, table 5). The second amounts to assuming the same self- assumption of 10 percent growth in yield would sufficiency ratio in 2025 as at present. give 3.176 million metric tons of productions from the existing irrigated and rain-fed land (row 10, • The yield in both irrigated and in rain-fed rice table 5). This amounts to 1,011 metric tons of in 2025 will be 10 percent greater than in 1991 additional rice production in 2025 (row 11, table 5).

TABLE 5. Growth in rice irrigated area from 1991 to 2025.

Year Row Factor Unit Irrigation Rain-fed Total 1991 1 Gross area 1,000 ha 575 261 836 2 % of total % 69 31 - 3 Yield mt/ha 3.84 2.60 - 4 Total production 1,000 mt 2,210 677 2,887 5 % of total % 77 23 100 2025 6 Population growth % - - 145 7 Rice demand 1,000 mt - - 4,187 8 Growth in yield % 10 10 - 9 Yield mt/ha 4.22 2.86 - 10 Production from 1991 area 1,000 mt 2,431 745 3,176 11 Total additional production requirement 1,000 mt 1,011 - 1,011 12 % additional production % 100 0 - 13 Additional production 1,000 mt 1,011 0 1,011 14 Additional required area 1,000 ha 239 0 241 15 Gross area 1,000 ha 814 261 1,077 16 Increase in area % 42 0 29

13 Under the last assumption, additional rice drinking water for survival, water for human production will have to be obtained only by hygiene, water for sanitation services, and water increasing the irrigated area. To obtain the for modest household needs for preparing food. He additional demand, the irrigated area has to be suggested a minimum level of 50 liters/person/day increased by 42 percent (row 16, table 5). for domestic use. The present per capita domestic The change in area—42 percent—from 1990 withdrawals of most districts are below this level. to 2025 is computed only for rice irrigation. Since We project the 2025 domestic needs at either the the production and yield data for OFCs are not BWR of 50 liters/person/day, as suggested by available, we assume that the OFC irrigated area Gleick, or the present withdrawal level, if the latter will also increase by 42 percent in 2025. In is higher. calculating the future demand, we ignored the fact We also assume at least 50 liters/person/day that that there is no potential for an increase in for 2025 per capita industrial demand. In the irrigated area in some districts, especially those in majority of the districts per capita industrial supply the wet zone. Also we ignored the possible losses in 1991 was well below this level. Only two of rain-fed area due to expansion in irrigated area. districts (Colombo and Gampaha) had withdrawn If these are considered, the required growth in more than 50 liters/person/day. The 2025 demand irrigated area in some districts may be higher than of these districts is projected at the 1991 per the value estimated here. capita level. Irrigation demand under the first scenario is estimated at 16 percent project efficiency level, or Total Demand 22 percent irrigated sector efficiency level. Under the second scenario, sector efficiency is assumed The total demand projections for the wet and dry to be 44 percent. zones are given in table 6. Because of its substantial share in irrigation, the dry zone Domestic and Industrial districts will require more than 80 percent of the total demand under both scenarios. At the same The demand projections for the domestic and time, because of high contribution to irrigation industrial sectors are based on meeting at least demand, the total water demand in Sri Lanka, the basic human requirement or keeping the especially in the dry zone, can be reduced by present level of per capita water withdrawals or almost one-half under the increased irrigation both. Gleick (1996) defined basic water efficiency scenario. requirement (BWR) for human needs in terms of:

TABLE 6. 2025 Demand projections.

Maha season Yala season D&I1 Irrigation Total D&I Irrigation Total Unit S12 S23 S1 S2 S2/S1 S1 S2 S1 S2 S2/S1 km3 km3 km3 km3 km3 %km3 km3 km3 km3 km3 % C14 C24 C34 C44 C54 C6 C74 C84 C94 C104 C114 C12 Sri Lanka 0.53 4.81 2.41 5.34 2.93 55% 0.53 5.18 2.59 5.71 3.12 55% Wet zone 62% 18% 18% 23% 26% 64% 62% 9% 9% 14% 18% 71% Dry zone 38% 82% 82% 77% 74% 52% 38% 91% 91% 86% 82% 52%

1D&I = Domestic and industrial. 2Scenario 1. 3Scenario 2. 4Wet-zone and dry-zone values are given as a percent of Sri Lankan total.

14 Water Scarcity Indicators: District-Wise Distribution

Water scarcity of a district can be due to low little or no water scarcity. The “MS” and “M” in the water supply or high demand or both. The per IWMI criteria indicate severe and moderate capita water resources, demand with respect to economic water scarcity, respectively. The lower supply, and the increase in demand are given in case letters: “b,” “m,” and “y” indicate whether the table 7. Table 8 shows the scarcity level of Sri indicated level of water scarcity is in both Lanka, two climatic zones, and the districts seasons, only in the maha season, or only in the according to the three criteria: Falkenmark, UN, yala season, respectively. For example “S-m” and IWMI. The letters “S,” “MS,” “M,” and “N,” indicates that the unit is severely water scarce in respectively, in table 8 indicate that a unit has the maha season but not in the yala season. either severe, medium to severe, moderate, or

TABLE 7. Per capita water resources and water withdrawals as a percent of water resources.

Per capita Maha season Yala season annual Withdrawal as % 2025 withdrawal - Withdrawal as % 2025 withdrawal - Unit water resources of water resources % change from of water resources % change from 1991 2025 1991 2025 1991 withdrawal 1991 2025 1991 withdrawal S1 S2 S1 S2 S1 S2 S1 S2 m3 m3 %%%%%%%%%% C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 Sri Lanka 2390 1645 16 23 13 49 -18 21 31 17 49 -19 Wet zone (WZ) 2513 1730 7 11 7 57 0 4 6 4 67 19 Dry zone (DZ) 2233 1537 24 35 18 47 -23 64 94 49 46 -24 Districts-WZ Colombo 586 403 17 24 21 45 30 13 19 18 45 36 Galle 3691 2541 1 2 2 206 187 0 1 1 223 209 Gampaha 1122 772 12 18 13 49 6 7 11 9 54 26 Kalutara 4569 3145 4 6 4 65 -1 2 4 2 77 15 Kandy 800 551 21 34 20 64 -3 10 19 13 89 27 Kegalle 5008 3446 3 5 3 73 9 1 2 2 91 31 Matara 2308 1589 16 24 13 52 -17 10 15 9 59 -8 Nuwara Eliya 2460 1693 11 17 10 53 -11 3 6 4 67 10 Ratnapura 6050 4164 5 8 5 54 -14 3 4 3 64 -2 Districts-DZ Ampara 3790 2608 28 41 21 44 -26 113 162 82 43 -27 Anuradhapura 1938 1334 55 80 41 46 -25 79 115 59 45 -26 Badulla 2697 1856 10 16 9 51 -18 22 34 19 53 -16 Batticaloa 2292 1577 11 17 9 49 -20 174 253 132 46 -24 Hambantota 2025 1394 57 83 43 45 -26 80 116 60 45 -26 Jaffna(1) 217 149 23 42 28 82 25 209 332 196 59 -6 Killinochchi 1031 710 102 146 75 44 -26 134 195 102 45 -24 Kurunegala 1041 717 46 68 36 49 -21 38 58 32 55 -14 Mannar 929 639 49 72 39 48 -21 22 34 20 58 -7 Matale 2770 1906 12 19 10 50 -19 18 27 15 54 -15 Monaragala 5562 3828 5 7 4 47 -20 14 21 12 50 -15 Mullaitivu 3144 2164 22 32 17 44 -26 76 110 58 46 -23 Polonnaruwa 9249 6366 19 27 14 44 -27 80 114 57 43 -28 Puttalam 538 370 71 107 58 49 -19 34 56 33 61 -3 Trincomalee 3080 2120 12 18 10 47 -22 94 135 69 44 -26 Vavuniya 2200 1514 53 76 39 44 -26 27 40 22 47 -21

15 TABLE 8. Levels of scarcity of different units.

Scarcity level1 according to the three indicators Unit Falkenmark UN2 IWMI3 1991 2025 1991 2025 2025 Scenario 1 Scenario 2 Scenario 1 Scenario 2 Sri Lanka N M MS-y MS-b M-b M-b N-b Wet zone (WZ) N N N-b M-m N-b M-b N-b Dry zone (DZ) N M S-y S-y S-y S-y N-b Districts-WZ Colombo MS S M-b MS-m MS-m M-b M-b Galle N N N-b N-b N-b MS-b MS-b Gampaha M MS M-m M-b M-m M-b M-y Kalutara N N N-b N-b N-b M-b N-b Kandy MS MS MS-m MS-m M-b M-b M-y Kegalle N N N-b N-b N-b M-b M-y Matara N M M-b MS-m M-m M-b N-b Nuwara Eliya N M M-m M-m N-b M-b N-b Ratnapura N N N-b N-b N-b M-b N-b

Districts-DZ

Ampara N N S-y S-b S-y S-y S-y Anuradhapura N M S-b S-b S-b S-b S-y Badulla N N MS-y MS-y M-y M-b N-b Batticaloa N M S-y S-y S-y S-y S-y Hambantota N M S-b S-b S-b S-b S-y Jaffna(1) S S S-y S-b S-y S-y S-y Killinochchi N MS S-b S-b S-b S-b S-b Kurunegala M MS S-m S-b MS-b S-b N-b Mannar MS MS S-m S-m SM-m S-m S-m Matale N N M-b MS-y M-y M-b N-b Monaragala N N M-y M-y M-y M-b N-b Mullaitivu N N S-y S-y S-y S-y S-y Polonnaruwa N N S-y S-y S-y S-y S-y Puttalam MS S S-m S-b S-m S-b S-m Trincomalee N N S-y S-y S-y S-y S-y Vavuniya N M S-m S-m SM-m S-m N-b

1N - Little or no water scarcity; M - Moderate water scarcity; MS - Medium to severe water scarcity; S - Severe water scarcity. 2b - Both seasons; m - maha season; y - yala season. 3MS, and M under IWMI criteria, respectively, indicate severe economic and moderate economic water scarcity.

16 Falkenmark Indicator If the current rate of irrigation sector efficiency continues into the future the water-scarcity picture Per capita water supply is the basis for the will become even worse. Under this scenario, the Falkenmark criterion. Both current and future per dry zone as a unit will have severe year-around capita water resources in Sri Lanka as a whole water scarcities. Some districts in the dry zone and also in the two climatic zones show no will have major problems in meeting future severe water scarcity (figure 4). However, five demand under the existing level of irrigation districts are already in the medium to severe efficiency. However, with the increased irrigation water-scarce categories. Three more will enter into efficiency scenario most of these districts can these categories by 2025. These 8 districts will meet their total 2025 demand by withdrawing water have 46 percent of the total population, but at or below the current level. Even under the contain only 25 percent of the land area and 22 increased irrigation efficiency scenario, four percent of the irrigated area. districts in the maha season and nine districts in the yala season are identified as severely water- scarce, as their withdrawals are still substantial UN Indicator percentages of the available water resources. Under the UN indicator, the wet zone as a unit The UN indicator addresses demand in terms of does not face serious water scarcity at present or available water resources. This indicator also in the future. However, some districts will have to shows no severe water scarcity at national level increase their current withdrawals substantially to at present or in the future. However, the dry zone meet the 2025 demand. in the yala season is already experiencing severe water scarcities. At district level, seven districts in the maha IWMI Indicator season (21% of the population, 35% of the land area, and 44% of the irrigated area) are in the The future demands with respect to available severe water-scarce category. Nine districts in the water resources and also relative to the current yala season (22% of the population, 43% of the demand are the basis for the IWMI water scarcity land area, and 63% of the irrigated area) are in criteria (Seckler et al. 1998; Seckler, Barker and this category (figure 5). Amarasinghe 1999). The majority of the 1991 population (61%) Under the first scenario, the 2025 water lived in areas where water withdrawals were less demand at national level is well below 50 percent than 20 percent of the water resources in each of the available water resources, but will require season. However, these areas contain only small an increase of about 49 percent over the 1991 percentages of the irrigated area (28% in maha withdrawal level. Thus at national level no severe and 19% in yala). The population in these areas (absolute or economic) water-scarce condition will depends on the food production from other exist by 2025. At the climatic zone level, the dry districts that are relatively more water-scarce. This zone in the yala season will be in the absolute dependency is rather important because the water-scarce condition. Some districts in the dry irrigation sector has a substantial proportion of the zone (7 in maha and 11 in yala) will face absolute present rice production. Therefore, any increase in water-scarce conditions in 2025 (figure 6). All the magnitude of water scarcities for the districts districts in the wet zone, except Galle, will have that are already water-scarce will have a no form of severe water scarcity. Galle district has substantial impact on the ability to meet the future ample water resources to develop its future food demand. demand. But this district will have year-around

17 AMPARA BATTICALOA MONARAGALA HAMBANTOTA TRINCOMALEE BADULLA POLONNARUWA MATALE KANDY NUWARA ELIYA NUWARA MATARA RATNAPURA VAVUNIYA MULLAITIVU ANURADHAPURA KEGALLE KILINOCHCHI GALLE KURUNEGALA 2025 MANNAR KALUTARA JAFFNA GAMPAHA COLOMBO PUTTALAM Severe Medium to severe Moderate Little or no Water scarcity 1 Water AMPARA BATTICALOA MONARAGALA HAMBANTOTA TRINCOMALEE BADULLA POLONNARUWA MATALE KANDY NUWARA ELIYA NUWARA MATARA RATNAPURA VAVUNIYA MULLAITIVU ANURADHAPURA KEGALLE KILINOCHCHI GALLE 1991 KURUNEGALA MANNAR KALUTARA JAFFNA GAMPAHA COLOMBO PUTTALAM FIGURE 4. Falkenmark indicator of water scarcity.

18 AMPARA AMPARA BATTICALOA BATTICALOA MONARAGALA MONARAGALA HAMBANTOTA HAMBANTOTA TRINCOMALEE TRINCOMALEE BADULLA BADULLA POLONNARUWA POLONNARUWA MATALE MATALE KANDY KANDY NUWARA ELIYA NUWARA ELIYA MATARA MATARA RATNAPURA RATNAPURA VAVUNIYA VAVUNIYA MULLAITIVU MULLAITIVU ANURADHAPURA ANURADHAPURA KEGALLE KEGALLE KILINOCHCHI KILINOCHCHI GALLE GALLE KURUNEGALA KURUNEGALA MANNAR MANNAR KALUTARA KALUTARA JAFFNA JAFFNA GAMPAHA GAMPAHA COLOMBO COLOMBO PUTTALAM PUTTALAM 2025 Scenario 2 AMPARA AMPARA BATTICALOA BATTICALOA MONARAGALA MONARAGALA HAMBANTOTA HAMBANTOTA TRINCOMALEE TRINCOMALEE BADULLA BADULLA POLONNARUWA POLONNARUWA MATALE MATALE KANDY KANDY NUWARA ELIYA NUWARA ELIYA MATARA MATARA RATNAPURA RATNAPURA VAVUNIYA VAVUNIYA MULLAITIVU MULLAITIVU ANURADHAPURA ANURADHAPURA KEGALLE KEGALLE KILINOCHCHI KILINOCHCHI GALLE GALLE KURUNEGALA KURUNEGALA MANNAR MANNAR KALUTARA KALUTARA JAFFNA JAFFNA GAMPAHA GAMPAHA COLOMBO COLOMBO PUTTALAM PUTTALAM 2025 Scenario 1 AMPARA AMPARA BATTICALOA BATTICALOA MONARAGALA MONARAGALA HAMBANTOTA HAMBANTOTA TRINCOMALEE TRINCOMALEE BADULLA BADULLA POLONNARUWA POLONNARUWA MATALE MATALE KANDY KANDY NUWARA ELIYA NUWARA ELIYA MATARA MATARA RATNAPURA RATNAPURA VAVUNIYA VAVUNIYA MULLAITIVU MULLAITIVU ANURADHAPURA ANURADHAPURA KEGALLE KEGALLE KILINOCHCHI KILINOCHCHI GALLE GALLE KURUNEGALA KURUNEGALA 1991 MANNAR MANNAR KALUTARA KALUTARA JAFFNA JAFFNA

GAMPAHA GAMPAHA COLOMBO COLOMBO PUTTALAM PUTTALAM

Maha season Maha season Yala Severe Medium to severe Moderate Little or no Water scarcity Water FIGURE 5. UN indicator of water scarcity.

19 FIGURE 6. IWMI indicator of water scarcity in 2025.

Scenario 1 Water scarcity Scenario 2

JAFFNA Severe absolute JAFFNA Severe economic

KILINOCHCHI Moderate economic KILINOCHCHI

MULLAITIVU Little or no MULLAITIVU

VAVUNIYA MANNAR VAVUNIYA MANNAR

ANURADHAPURA TRINCOMALEE ANURADHAPURA TRINCOMALEE

PUTTALAM POLONNARUWA PUTTALAM POLONNARUWA

BATTICALOA BATTICALOA

KURUNEGALA MATALE KURUNEGALA MATALE Maha season

AMPARA KANDY AMPARA KANDY GAMPAHA KEGALLE GAMPAHA KEGALLE NUWARA ELIYA NUWARA ELIYA BADULLA COLOMBO BADULLA COLOMBO MONARAGALA MONARAGALA RATNAPURA KALUTARA RATNAPURA KALUTARA

HAMBANTOTA GALLE HAMBANTOTA MATARA GALLE MATARA

JAFFNA JAFFNA

KILINOCHCHI KILINOCHCHI

MULLAITIVU MULLAITIVU

VAVUNIYA MANNAR VAVUNIYA MANNAR

ANURADHAPURA TRINCOMALEE ANURADHAPURA TRINCOMALEE

PUTTALAM POLONNARUWA PUTTALAM POLONNARUWA

BATTICALOA BATTICALOA

KURUNEGALA MATALE KURUNEGALA MATALE Yala season Yala

AMPARA KANDY AMPARA KANDY GAMPAHA KEGALLE GAMPAHA KEGALLE NUWARA ELIYA NUWARA ELIYA BADULLA COLOMBO BADULLA COLOMBO MONARAGALA MONARAGALA RATNAPURA KALUTARA RATNAPURA KALUTARA

HAMBANTOTA GALLE HAMBANTOTA MATARA GALLE MATARA

20 severe economic water scarcity as it will have to availability areas within the country or from triple its 1991 withdrawal levels to meet the 2025 outside the country. demand. On the other hand, the high per capita water Under the increased efficiency scenario, only supply does not necessarily indicate the two districts (4% of the population) in the maha possibilities for further water resources season, and nine districts (22% of the population) developments. Both UN and IWMI indicators show in the yala season are in the absolute water-scarce that the demand with respect to water availability category. These districts are located in the dry is high in some places where per capita water zone. Under this scenario Galle district will still be supply is very high. in severe economic water-scarce conditions. For example, Ampara, Polonnaruwa, Trincomalee, and Mullaitive districts have very high annual per capita water resources. Under the Comparisons of the Three Indicators same irrigation project efficiency scenario, these districts will face at least severe seasonal water None of the water scarcity criteria indicate severe scarcities according to UN and IWMI indicators. seasonal or year-round water scarcity at national Therefore, for meeting future demand, they will level (table 8). However, there are several districts have to increase their irrigation sector efficiencies that are experiencing serious water scarcities at or import water from other districts. present or will enter into this category by the year Some of the districts may have high per capita 2025. annual water resources and also low levels of The Falkenmark indicator shows that almost current withdrawals compared to available water half of the 2025 population will be in districts resources. These districts are not identified as where per capita water availability is very low. water scarce by the Falkenmark and UN indicators. This level of low water availability will be a severe But the IWMI indicator identifies Galle district as constraint for further water resources development, severely water-scarce in the economic sense, especially for agriculture. Increases in food which means that it will have to develop a requirement in these areas will have to be met substantial amount of withdrawals to meet with imports either from the high per capita water agricultural, domestic, and industrial needs in 2025.

Conclusion

Results of the study show that aggregated will be severely water-scarce in the economic national-level statistics on water scarcity are sense so that they will have to develop a indeed misleading even for a small country like Sri substantial amount of withdrawals over the present Lanka in monsoonal Asia. National level statistics level to meet their 2025 needs. often ignore wide temporal and spatial variations It should be noted here that all districts in the of water availability and demand. Despite what wet zone have a smaller share of the country’s national level statistics say, water availability in irrigation withdrawal compared to the population. some districts will be a significant constraint for They will require more domestic and industrial future social and economic development. Also, withdrawals, and will depend heavily on the contrary to the common belief, some districts in irrigated rice production from other districts. the wet zone with ample per capita water supply However, the major irrigated rice producing

21 districts are already using a substantial amount of percent. Under this scenario the savings of water their available water resources. in the agriculture sector are more than adequate to If the current rate of irrigation efficiency meet the future additional demands in the other continues in the future, several districts with major two sectors. rice irrigated areas will have severe water Whether the country has the financial and scarcities according to UN and IWMI criteria. institutional capacity to attain high irrigation Some of these districts will be in serious water- efficiency in the second scenario is not clear. scarce conditions so that the available water However, at the current irrigation efficiency level, resources may not be adequate even to meet their the majority of dry zone districts will face either projected demand. These districts will have to seasonal or year-round severe water scarcities. either reduce demand for irrigation by increasing Also the costs and benefits of increasing irrigation efficiency or by importing food, or efficiency versus importing more food, or transport water from regions in the wet zone where transporting water from water-abundant to water- water is abundant. scarce areas are not studied in this report. The increased irrigation sector efficiency Because of current water scarcities in the major scenario shows that, by doubling the irrigation rice irrigating districts, these are important areas efficiency, the total demand in 2025 of most of research that are useful for future policy water-scarce districts can be reduced by 50 planning.

22 APPENDIX A.

Runoff Estimates of River Basins in Sri Lanka

For the current study, we estimate the primary adjoining basins. The average and the 75 percent discharges (runoff) to the sea from the river runoff estimates for the maha and yala seasons, basins at 75 percent exceedence probability and the annual runoff at 75 percent and 50 rainfall level. Long-Term Hydrometeorological Data percent exceedence probability level rainfall and in Sri Lanka (Nakagawa et al. 1995) is the data also average rainfall are given in appendix table A. source for this analysis. There are 103 distinct For the purpose of comparison, the annual natural river basins6 in Sri Lanka covering 90 average runoff in the National Atlas of Sri Lanka percent of the total land area. Of these, only a (Survey Department of Sri Lanka 1988) is given in few river basins contain long-term monthly river the last column. The river basins are ordered discharge estimates. The period from 1945 to anticlockwise starting from the Kelani basin, as 1970 was considered for estimating the rainfall- indicated in the Sri Lankan Atlas. runoff relationships. Here we assumed that no Appendix table A shows that the average major development works were undertaken during runoff values we estimated are almost equal to this period. However, a few basins do contain the average values given in the National Atlas of some storage capacity constructed before 1945. Sri Lanka (Survey Department of Sri Lanka 1988). In these basins, the estimated seasonal runoff The average runoff estimate of most basins is was revised by adding the possible storage at the more than the runoff estimated at 50 percent end of the season. exceedence probability (median) rainfall level. This First, we estimate the seasonal (maha and implies that in the long run the estimated average yala) rainfall-runoff relationships for these basins. runoff cannot be expected even in 2 out of every 4 From the estimated relationship, we estimate the years. However, the runoff at 75 percent river discharge at 75 percent exceedence exceedence probability rainfall can be expected in probability basin rainfall. For the basins where at least 3 out of every 4 years in the long run. data are not available, we linearly interpolate the Therefore, the seasonal runoff at 75 percent runoff estimates per unit area (mm) of the two exceedence probability rainfall is used in this report.

6In addition, there are 94 coastal basins with no significant contribution to freshwater resources (Arumugam 1969).

23 APPENDIX TABLE A. Runoff estimates of river basins of Sri Lanka.

No. Name of river Drainage Seasonal runoff Annual runoff basin area P75 Estimate Sri Lankan Maha Yala P75 P50 Average Atlas average km2 km3 km3 km3 km3 km3 km3

1 Kelani Ganga 2,278 2.33 2.97 5.30 5.63 5.75 5.47 2 Bolgoda Lake 374 0.41 0.50 0.92 0.99 1.02 0.75 3 Kaluganga 2,688 3.17 3.70 6.87 7.58 7.86 7.86 4 Bentota Ganga 760 0.73 0.85 1.58 1.75 1.81 1.54 5 Madu Ganga 59 0.07 0.08 0.15 0.17 0.17 0.15 6 Madampe Lake 90 0.10 0.12 0.23 0.25 0.26 0.17 7 Telwatte Ganga 51 0.06 0.07 0.13 0.14 0.15 0.10 8 Ratgama Lake 10 0.01 0.01 0.02 0.03 0.03 0.02 9 Gin Ganga 922 1.04 1.26 2.30 2.56 2.62 1.90 10 Koggala Lake 64 0.06 0.07 0.14 0.16 0.16 0.06 11 Polwatta Ganga 233 0.21 0.20 0.41 0.48 0.49 0.12 12 Nilwala Ganga 960 0.73 0.61 1.34 1.63 1.66 1.10 13 Sinimodara Oya 38 0.03 0.02 0.05 0.06 0.06 0.02 14 Kirama Oya 223 0.14 0.11 0.26 0.32 0.32 0.12 15 Rekawa Oya 92 0.04 0.03 0.08 0.10 0.10 0.04 16 Urubokke Oya 348 0.18 0.14 0.32 0.39 0.40 0.20 17 Kachigala Ara 220 0.10 0.07 0.18 0.22 0.22 0.06 18 Walawe Ganga 2,442 0.98 0.66 1.65 2.06 2.14 2.17 19 Karagan Oya 58 0.02 0.01 0.03 0.04 0.04 0.02 20 Malala Oya 399 0.11 0.08 0.19 0.24 0.25 0.07 21 Embilikala Oya 59 0.01 0.01 0.02 0.03 0.03 0.02 22 Kirindi Oya 1,165 0.20 0.15 0.35 0.43 0.46 0.48 23 Bambawe Ara 79 0.01 0.01 0.02 0.03 0.03 0.03 24 Mahasilawa Oya 13 0.00 0.00 0.00 0.00 0.00 0.01 25 Butawa Oya 38 0.01 0.00 0.01 0.01 0.01 0.02 26 Menik Ganga 1,272 0.17 0.08 0.25 0.32 0.30 0.49 27 Katupila Aru 86 0.01 0.00 0.02 0.02 0.02 0.03 28 Kuranda Ara 131 0.02 0.01 0.03 0.04 0.04 0.05 29 Namadagas Ara 46 0.02 0.01 0.03 0.04 0.04 0.04 30 Kambe Ara 46 0.01 0.00 0.01 0.02 0.02 0.02 31 Kumbukkan Oya 1,218 0.30 0.05 0.36 0.47 0.48 0.77 32 Bangura Oya 92 0.02 0.00 0.03 0.04 0.04 0.04 33 Girikula Oya 15 0.00 0.00 0.01 0.01 0.01 0.01 34 Heawa Ara 51 0.02 0.00 0.02 0.02 0.02 0.03 35 Wila Ara 484 0.15 0.02 0.17 0.22 0.24 0.22 36 Heda Oya 604 0.23 0.02 0.25 0.34 0.36 0.39 37 Karanda Oya 422 0.17 0.01 0.18 0.24 0.26 0.20 38 Simena Ara 51 0.02 0.00 0.02 0.03 0.03 0.03 39 Tandiadi Aru 22 0.01 0.00 0.01 0.01 0.01 0.02 40 Kangikadichi Ara 56 0.02 0.00 0.03 0.03 0.04 0.04 41 Rufus Kulam 35 0.01 0.00 0.02 0.02 0.02 0.02 42 Pannel Oya 184 0.08 0.00 0.09 0.11 0.12 0.13 43 Ambalama Oya 115 0.05 0.00 0.05 0.07 0.08 0.08 44 Gal Oya 1,792 0.82 0.03 0.86 1.11 1.26 1.25 45 Andella Oya 522 0.24 0.01 0.25 0.33 0.37 0.29 46 Thumpankeni Tank 9 0.00 0.00 0.00 0.01 0.01 0.01 47 Namakada Aru 12 0.01 0.00 0.01 0.01 0.01 0.01 48 Mandipattu Aru 100 0.05 0.00 0.05 0.07 0.08 0.09 49 Pattanthe Dephue Aru 100 0.01 0.00 0.01 0.02 0.02 0.09 50 Magalatavan Aru 346 0.18 0.00 0.18 0.23 0.27 0.29 (Continued)

24 No. Name of river Drainage Seasonal runoff Annual runoff basin area P75 Estimate Sri Lankan Maha Yala P75 P50 Average Atlas average km2 km3 km3 km3 km3 km3 km3 51 Vett Aru 26 0.01 0.00 0.01 0.02 0.02 0.02 52 Mundeni Aru 1,280 0.58 0.02 0.60 0.77 0.88 0.86 53 Miyangollal Ela 225 0.09 0.00 0.09 0.12 0.13 0.12 54 Maduru Oya 1,541 0.50 0.04 0.54 0.75 0.75 0.81 55 Pulliyanpotha Aru 52 0.02 0.00 0.02 0.03 0.03 0.03 56 Kirimechi Odai 77 0.03 0.01 0.04 0.05 0.05 0.04 57 Bodigoda Aru 164 0.07 0.02 0.09 0.11 0.12 0.08 58 Mandan Aru 13 0.01 0.00 0.01 0.01 0.01 0.01 59 Makarachchi Aru 37 0.02 0.01 0.03 0.03 0.03 0.02 60 Mahaweli Ganga 10,327 5.37 2.71 8.08 9.09 9.72 11.02 61 Kantalai Basin Per Ara 445 0.13 0.01 0.14 0.20 0.20 0.15 62 Panna Oya 69 0.02 0.00 0.02 0.03 0.03 0.03 63 Palampotta Aru 143 0.03 0.00 0.04 0.05 0.06 0.06 64 Pankulam Ara 382 0.08 0.01 0.09 0.13 0.14 0.17 65 Kanchikamban Aru 205 0.04 0.00 0.04 0.06 0.07 0.08 66 Palakutti Aru 20 0.00 0.00 0.00 0.01 0.01 0.01 67 Yan Oya 1,520 0.20 0.02 0.22 0.36 0.38 0.30 68 Mee Oya 90 0.01 0.00 0.01 0.02 0.02 0.03 69 Ma Oya 1,024 0.13 0.01 0.15 0.24 0.26 0.31 70 Churian Aru 74 0.01 0.00 0.01 0.02 0.02 0.03 71 Chavar Aru 31 0.00 0.00 0.00 0.01 0.01 0.01 72 Palladi Aru 61 0.01 0.00 0.01 0.01 0.02 0.02 73 Nay ARa 187 0.02 0.00 0.03 0.04 0.05 0.07 74 Kodalikallu Aru 74 0.01 0.00 0.01 0.02 0.02 0.03 75 Per Ara 374 0.05 0.01 0.05 0.08 0.09 0.12 76 Pali Aru 84 0.01 0.00 0.01 0.02 0.02 0.03 77 Muruthapilly Aru 41 0.00 0.00 0.01 0.01 0.01 0.02 78 Thoravil Aru 90 0.01 0.00 0.01 0.02 0.02 0.03 79 Piramenthal Aru 82 0.01 0.00 0.01 0.02 0.02 0.03 80 Nethali Aru 120 0.01 0.00 0.02 0.02 0.03 0.03 81 Kanakarayan Aru 896 0.10 0.02 0.12 0.18 0.22 0.02 82 Kalawalappu Aru 56 0.01 0.00 0.01 0.01 0.01 0.01 83 Akkarayan Aru 192 0.02 0.00 0.02 0.04 0.05 0.04 84 Mendekal Aru 297 0.03 0.01 0.04 0.06 0.07 0.08 85 Pallarayan Kadu 159 0.02 0.00 0.02 0.03 0.04 0.05 86 Pali Aru 451 0.05 0.01 0.06 0.08 0.11 0.11 87 Chappi Aru 66 0.01 0.00 0.01 0.01 0.02 0.02 88 Parangi Aru 832 0.08 0.02 0.10 0.15 0.20 0.27 89 Nay Aru 560 0.06 0.01 0.07 0.10 0.13 0.12 90 Malvatu Oya 3,246 0.32 0.08 0.40 0.58 0.77 0.57 91 Kal Ara 210 0.02 0.01 0.02 0.04 0.05 0.04 92 Moderagam Ara 932 0.08 0.03 0.11 0.17 0.23 0.16 93 Kala Oya 2,772 0.23 0.08 0.31 0.50 0.68 0.59 94 Moongil Aru 44 0.00 0.00 0.00 0.01 0.01 0.01 95 Mi Oya 1,516 0.12 0.01 0.13 0.21 0.26 0.34 96 Madurankuli Aru 62 0.01 0.00 0.01 0.01 0.01 0.03 97 Kalagamuwa Oya 151 0.02 0.01 0.03 0.04 0.05 0.05 98 Pantampola Oya 215 0.04 0.02 0.05 0.07 0.08 0.07 99 Deduru Oya 2,616 0.52 0.26 0.78 0.98 1.13 1.61 100 Karambala Oya 589 0.17 0.12 0.29 0.35 0.37 0.25 101 Ratmal Oya 215 0.08 0.07 0.15 0.17 0.18 0.09 102 Maha Oya 1,510 0.69 0.61 1.31 1.53 1.54 1.61 103 Attanagalu Oya 727 0.54 0.62 1.16 1.27 1.29 1.57 Total 59,671 24.14 16.91 41.05 48.04 50.89 50.59

25 APPENDIX B.

Irrigation Requirement

The irrigation requirement for each district is given The NET for the maha season is the sum of in appendix table B. The net evapotranspiration the NETs for the five months from October to (NET), for a period of 1 month is defined as the February. The NET for the yala season is the sum evapotranspiration for the reference crop (ETo) of the NETs for the four months from April to July. minus the effective rainfall. The effective rainfall is The monthly ETo and average rainfall data for a fraction of the average rainfall. To calculate this each district were obtained from the Climate and fraction, first the total average rainfall volume and Water Atlas (IIMI and Utah State University 1997). the runoff volume generated from this rainfall on a The computation of NET on irrigated area is unit are calculated. The difference between the shown in appendix table B. The rice and OFC- rainfall and runoff volumes is the fraction of irrigated areas in the 1991 maha season are in C1 rainfall that is available at the place of rainfall and C2, respectively. The 1991 yala season data occurrence. We further assume that 10 percent of are in columns C7 and C8. The irrigated rice area this difference is lost as deep percolation, and the data were obtained from the Department of remaining 90 percent as the effective rainfall is Census and Statistics 1995, and the irrigated OFC available for crop use. As mentioned by Seckler area data were obtained from Jayawardana, et al. (1998), in the case of rice we multiply the Jayasinghe, and Dayarathne 1993. The NETs for ETo by 110 percent to obtain NET for rice. This is the maha season rice and OFC are given in C3 because most rice fields are flooded during land and C4, and the NETs for the yala season preparation and the flooded water surface is prone irrigated rice and OFC are in C8 and C9. to higher evaporation. If the difference between The total NETs (C6, C12) on the irrigated area ETo and effective rainfall is negative, NET for the for maha and yala seasons as depths are given in month is taken to be zero, i.e., there is no C5 and C11. irrigation requirement for that month.

26 APPENDIX TABLE B.

Maha season, 1991 Yala season, 1991 Irrigated area NET Irrigated area NET Unit Paddy OFC Paddy OFC on irr. area Paddy OFC Paddy OFC on irr. area Depth Total depth total 1,000ha 1,000ha m m m km3 1,000ha 1,000ha m m m km3 C11 C21 C3 C4 C5 C61 C71 C81 C9 C10 C11 C121 Sri Lanka 347.8 28.3 0.23 0.19 0.19 0.70 227.29 38.87 0.30 0.25 0.28 0.76 Wet Zone (WZ) 13.5% 0.3% 0.33 0.28 0.27 18.4% 16.4% 3.2% 0.22 0.17 0.17 8.8% Dry Zone (DZ) 86.5% 99.7% 0.17 0.14 0.17 81.6% 83.6% 96.8% 0.35 0.30 0.30 91.2% Districts-WZ Colombo 0.2% 0.45 0.39 0.45 0.5% 0.3% 0.31 0.27 0.31 0.3% Galle 0.0% 0.46 0.40 0.46 0.1% 0.1% 0.30 0.25 0.30 0.0% Gampaha 1.1% 0.34 0.28 0.34 1.8% 1.1% 0.20 0.15 0.20 0.7% Kalutara 0.9% 0.43 0.38 0.43 2.0% 1.3% 0.29 0.25 0.29 1.2% Kandy 3.5% 0.17 0.14 0.17 3.0% 4.3% 2.4% 0.09 0.04 0.08 1.2% Kegalle 0.7% 0.32 0.26 0.32 1.2% 1.1% 0.19 0.15 0.19 0.6% Matara 2.3% 0.37 0.32 0.37 4.2% 2.8% 0.26 0.21 0.26 2.2% Nuwara Eliya 1.7% 0.18 0.15 0.18 1.5% 1.4% 0.14 0.10 0.14 0.6% Ratnapura 2.9% 0.3% 0.28 0.23 0.28 4.0% 4.0% 0.8% 0.17 0.12 0.16 2.0% Districts-DZ Ampara 11.1% 0.1% 0.19 0.15 0.19 10.4% 19.8% 2.9% 0.29 0.24 0.29 17.8% Anuradhapura 15.5% 1.8% 0.16 0.14 0.16 12.3% 6.8% 39.6% 0.35 0.30 0.33 13.4% Badulla 5.3% 0.5% 0.15 0.13 0.15 4.1% 4.5% 4.4% 0.22 0.17 0.21 3.4% Batticaloa 3.5% 8.0% 0.14 0.10 0.13 2.7% 4.6% 0.37 0.32 0.37 5.2% Hambantota 7.8% 16.9% 0.23 0.19 0.22 10.2% 10.3% 3.9% 0.30 0.25 0.29 9.6% Jaffna(1) 0.0% 15.3% 0.15 0.13 0.13 0.8% 0.0% 8.7% 0.51 0.46 0.46 2.0% Killinochchi 2.6% 4.2% 0.15 0.13 0.15 2.1% 0.4% 2.5% 0.50 0.44 0.47 1.2% Kurunegala 11.5% 0.6% 0.21 0.17 0.21 11.9% 9.8% 23.8% 0.16 0.11 0.14 6.0% Mannar 0.7% 5.6% 0.17 0.15 0.16 0.9% 0.2% 0.42 0.36 0.42 0.3% Matale 3.2% 0.4% 0.17 0.15 0.17 2.8% 1.7% 3.4% 0.28 0.23 0.27 1.8% Monaragala 2.9% 0.1% 0.14 0.12 0.14 2.0% 1.8% 0.6% 0.22 0.17 0.22 1.2% Mullaitivu 1.5% 13.4% 0.14 0.12 0.13 1.6% 0.6% 0.48 0.43 0.48 0.8% Polonnaruwa 12.1% 2.7% 0.18 0.15 0.18 11.1% 16.8% 0.41 0.36 0.41 20.8% Puttalam 2.8% 8.1% 0.22 0.18 0.21 3.7% 2.0% 4.2% 0.16 0.12 0.15 1.2% Trincomalee 4.4% 0.0% 0.12 0.10 0.12 2.7% 4.2% 0.45 0.40 0.45 5.8% Vavuniya 1.6% 22.0% 0.15 0.13 0.14 2.3% 0.1% 2.9% 0.44 0.38 0.39 0.7% 1Wet-zone, dry-zone, and district values are given as a percent of the Sri Lankan total.

27 Literature Cited

Aluwihare, P. B.; and M. Kikuchi. 1991. Irrigation investment trends in Sri Lanka: New construction and beyond. Co- lombo, Sri Lanka: International Irrigation Management Institute.

Arumugam, S. 1969. Water resources of Ceylon: Its utilization and development. Colombo, Sri Lanka: Water Re- sources Board.

Department of Census and Statistics. 1995. Statistical abstract of the Democratic Socialist Republic of Sri Lanka 1994. Colombo, Sri Lanka: Department of Census and Statistics. Engelman, R.; and P. Leroy. 1993. Sustaining water: Population and the future of renewable water supplies. Popu- lation and Environment Program. Washington, D.C.: Population Action International. Economic and Social Commission for Asia and the Pacific (United Nations). 1995. Guidebook to water resources, use and management in Asia and the Pacific. Volume one: Water resources and water use. Water resources se- ries no.74. New York, NY, USA: United Nations. Falkenmark, Malin; Jan Lundqvist; and Carl Widstrand. 1989. Macro-scale water scarcity requires micro-scale ap- proaches: Aspects of vulnerability in semi-arid development. Natural Resources Forum 13 (4): 258-267. Gleick, Peter H., ed. 1996. Basic water requirements for human activities: Meeting basic needs. Water International 21(2): 83-92. International Irrigation Management Institute (IIMI) and Utah State University. 1997. World water and climate atlas. Colombo, Sri Lanka: International Irrigation Management Institute. Irrigation Department. 1975. Register of irrigation projects in Sri Lanka. Colombo, Sri Lanka: Irrigation Department.

Jayawardana, J.; Ananda Jayasinghe; and P.W.C. Dayarathne. 1993. Promoting implementation of crop diversifica- tion in rice-based irrigation systems in Sri Lanka. In Promoting crop diversification in rice-based irrigation systems, ed. Senen M. Miranda and R. Maglinao. Colombo, Sri Lanka: International Irrigation Management Institute.

Mahaweli Authority of Sri Lanka. 1995. Mahaweli statistical handbook 1994. Colombo, Sri Lanka: Mahaweli Author- ity of Sri Lanka.

Nakagawa, K.; H. Edagawa; V. Nandakumar; and M. Aoki. 1995. Long-term hydrometeorological data in Sri Lanka: Data book of hydrological cycles in humid tropical ecosystems, Part I. Japan: University of Tsukuba. Raskin, P.; P. Gleick; P. Kirshen; G. Pontius; and K. Strzepek. 1997. Water futures: Assessment of long-range patterns and problems. Stockholm, Sweden: Stockholm Environment Institute. Sakthivadivel, R.; C. R. Panabokke; C. M. Wijayaratna; N. Fernando; K. Jinapala; and R. B. Bandula Sirimal. 1995. Natural resources management study of North Central Province Sri Lanka. Annexures - Volume II: Pre-project tech- nical assistance study for proposed Area Development Project of North Central Province. Colombo: International Irrigation Management Institute.

Samad, M. 1997. Personal communication. Seckler, D.; Upali. A. Amarasinghe; David Molden; Radhika de Silva; and Randy Baker. 1998. World water demand and supply 1990 to 2025: Scenarios and issues. Research Report 19. Colombo: International Irrigation Manage- ment Institute. Seckler, D.; Randy Barker; and Upali A. Amarasinghe. 1999. Water scarcity in the twenty-first century. International Journal of Water Resources Development 15(1/2):29-42. Survey Department of Sri Lanka. 1988. The national atlas of Sri Lanka. Sri Lanka: Survey Department of Sri Lanka.

United Nations. 1995. World population prospects: The 1994 revision. New York, NY, USA: United Nations.

28 Wijayaratna, C. M.; and K. Hemakeerthi. 1992. Production and profitability of rice cultivation in Sri Lanka under dif- ferent water regimes: A trend analysis. In Advancements in IIMI’s research 1992: A selection of papers presented at the Internal Program Review, ed. International Irrigation Management Institute, 277-308. Colombo, Sri Lanka: International Irrigation Management Institute.

29 Research Reports

19. World Water Demand and Supply, 1990 to 2025: Scenarios and Issues. David Seckler, Upali Amarasinghe, David Molden, Rhadika de Silva, and Randolph Barker, 1998.

20. Indicators for Comparing Performance of Irrigated Agricultural Systems. David J. Molden, R. Sakthivadivel, Christopher J. Perry, Charlotte de Fraiture, and Wim H. Kloezen, 1998.

21. Need for Institutional Impact Assessment in Planning Irrigation System Modernization. D. J. Bandaragoda, 1998.

22. Assessing Irrigation Performance with Comparative Indicators: The Case of the Alto Rio Lerma Irrigation District, Mexico. Wim H. Kloezen and Carlos Garcés-Restrepo, 1998.

23. Performance of Two Transferred Modules in the Lagunera Region: Water Relations. G. Levine, A. Cruz, D. Garcia, C. Garcés-Restrepo, and S. Johnson III, 1998.

24. Farmer Response to Rationed and Uncertain Irrigation Supplies. C. J. Perry and S. G. Narayanamurthy, 1998.

25. Impacts of Colombia’s Current Irrigation Management Transfer Program. Douglas L. Vermillion and Carlos Garcés-Restrepo, 1998.

26. Use of Historical Data as a Decision Support Tool in Watershed Management: A Case Study of the Upper Nilwala Basin in Sri Lanka. W. K. B. Elkaduwa and R. Sakthivadivel, 1998

27. Performance Evaluation of the Bhakra Irrigation System, India, Using Advanced Information Technologies. Wim Bastiaanssen and D. Molden, 1998.

28. Performance Evaluation of the Bhakra Irrigation System, India, Using Remote Sensing and GIS Techniques. R. Sakthivadivel, S. Thiruvengadachari, Upali Amerasinghe, W.G.M. Bastiaanssen, and David Molden, 1999.

29. Generic Typology for Irrigation Systems Operation. D. Renault, and G.G.A. Godaliyadda, 1999.

30. Mechanically Reclaiming Abandoned Saline Soils: A Numerical Evaluation. S. A. Prathapar and Asad S. Qureshi, 1999.

31. Gender Issues and Women’s Participation in Irrigated Agriculture: The Case of Two Private Irrigation Canals in Carchi, Ecuador. Elena P. Bastidas, 1999.

32. Water Scarcity Variations within a Country: A Case Study of Sri Lanka. Upali A. Amarasinghe, Lal Mutuwatta, and R. Sakthivadivel, 1999. 32 Research Report

Water Scarcity Variations within a Country: A Case Study of Sri Lanka

Upali A. Amarasinghe Lal Mutuwatta and R. Sakthivadivel

INTERNATIONAL WATER MANAGEMENT INSTITUTE P O Box 2075, Colombo, Sri Lanka Tel (94-1) 867404 • Fax (94-1) 866854 • E-mail [email protected] Internet Home Page http://www.cgiar.org/iwmi ISBN 92-9090-383-X ISSN 1026-0862 International Water Management Institute